Quantitative Analysis of Civ Traits

[vmxa1]

whoops forgot my .02 on leaders/military. I had three as Persia and did not even build the EPIC or I would have had more.

[mharmless]

Also I think building the heroic epic counters most of the militaristic leader advantages (now I admit this doesn't happen every game.)

Heroic Epic definitly stacks with millitaristic, not replaces it. Two games now I've had eight leaders, both were with epic+militant. The most I've gotten as a non militant was four (with epic too), in a game where I was playing like I do as a militant anyway.

I'd argue that Religion is more valuable than military for a couple of reasons. The ability to cheaply build a temple in a conquored cities (which allows you to expand your culture/border to 2 squares) is very helpful. You can also do this with scientific (but generally you need the temple and the libary is kinda of useless).

The single bigest factor I've seen in holding captured cities is population. I whip them down to nothing as fast as I can. Usualy I can get temple + barracks + cathedral without too much difficulty, then I let the town grow back up. The new citizens are all roman, no problems for me anymore. It takes 2 citizens instead of 1 if you aren't religious... GOOD. I want to kill 'em off quickly anyway.

I personaly would rank militant as one of the top abilities, but no ammount of yaking is going to convince anybody else. Try it. Play a militant. I played my first games as greek because they look so good on paper. I played my second batch of games as the french, again because they look good on paper. After that, a serries of randomly assigned civs. I ended up with the romans twice, and they just fit my playstyle so well compared to all those non-militants I tried first that I most often choose to play them now.

[vmxa1]

I think they can all work, except expanist. I do not see what it offers. Maybe someone can expalin it to me, other than a scout to start.

[Badtz Maru]

Originally posted by vmxa1
I think they can all work, except expanist. I do not see what it offers. Maybe someone can expalin it to me, other than a scout to start.

Expansionist civilizations also get better stuff from huts, and never anything bad from them. It's a lot more useful on larger maps, I usually play on Huge so Expansionist gives me a great advantage early on. I get at least a couple of techs from huts, usually a military unit or two, and quite often an extra settler. You also get a bit of money in your treasury early on, which gives you a lot more freedom in your budget. You will encounter the other civilizations early and thus be able to start trading for other techs, and you will know where to build your 'border cities' so that you can claim lots of terrain and block AI expansion (I don't see leaving spaces in your empire as a problem, as the cities the AI inevitably builds in them will always end up being yours in the long-run). On a Huge map with Expansionist I almost always have the largest empire as far as land area and I establish a tech lead very early on. A good start can make up for not having any benefit later in the game.

[Grim Legacy]

D'oh been playing too long to think straight. Greek. Greek.

[vmxa1]

I have a huge game going now and I made one warrior and 3-4 hoplites and was able to get to nearly all of the huts first. I say this as I found huts very far from my base (30+ turns to get back) and within 3 turns of their cities, so I am not convinced that the scout is required. I still say it is the weakest link.

[David Weldon]

I've started a thread on inflation so that delayed benefits can be discounted to the beginning of the game. This will help complicate your math quite a bit!

It's not fleshed out yet, but it would be great if anyone that wants to could
contribute or read here.

[JMan]

I agree that it's not enough to say that one Industrious worker is equal to two non-Industrious workers. The same is true for Militaristic military units and the way they were related to non-Militaristic military units - the difference is more than 20% or however it was expressed.

Forget upkeep costs for a moment - they are important, but only under certain governments. The real difference is the time that building two workers would take instead of building one. At minimum, the difference will be at least one city-turn. (You know, city-turn, like man-hour). In that turn, the city could be building something else instead. I believe that economists refer to this as "opportunity cost". It's exactly the same for military units - non-Militaristic civs will have to build more of them to get the same effect.

I suppose you could say that the opportunity costs would cancel out, with Religious civs, for instance, saving opportunity cost on Temples, Cathedrals, etc. However, that seems unsatisfactory to me.

[ThaddeusAlexander]

Yay! persians 4 Evr !!

Thanks for all of your hard work and thought ...

~Alex

[B-line]

...this relative worth all depends on the map type you are playing, and the difficulty setting. The game empthasis changes dramatically when you move off the standard world map.

At the hardest settings city moral is extremely important and hard to maintain, and corruption ins the early game can be crippling.

With raging barbarians selected, strong defence is needed very early from large stacks of mounted barbarians.

In a small map, technologies are cheeper, exploration faster and the number of cities you will end up with (hence corruption) greatly reduced.

In an archepelico map, opportunities for military conquest are generally post-poned till sea travell is well established and defence can be scalled down.

With one large contentent, military strength can be very pivatol.


I have found the following holds for small island maps...

Expansionist: next to useless. Great for newbies at low difficulties and big maps, but just doesn't cut it at the harder levels.

Militaristic: Handy for geating leaders early.. Dominating the better World wonders construction early always leads to an easy win, also dominating another player early can be very satisfying (and rewarding) but If you start on your own little island then you will get little benifit from this skill.

Religion: Cheap temples and cathedrals are worth their weight in gold at the higher difficulty settings. Typically a towns production will be greatly curtailed due to too many dicontented citizens. To compensate you will either up luxuries (loosing research), building more warriers to squat in each town...waisting precious time, or turning off production (very painfull)
Of cause I always rush builds temples and Cathedrals as soon as I can do it for just one population point. With this skill I sometimes revert back to depotism so I can rush build for my newer towns. Also as you build these items quickers, your culture grows earlier. I have found this skill to be the best "early game" skill available.

Industrialist: Faster workers sounds great, but in reality, in the early game my one non-industrial worker per town keeps up the towns growth demand, wonce you hit size six and can grow because of the lack of aquiducts then its cheep and fast to pump out a few more. by the time I get round to clearing those cursed jungles I usually have so many workers spare I can stack them up and clear each square in one or two turns. The extra production point for large cities is only valuable in the late game, where it is less valuable. ... Although it can mean the diffenence between building that essential wonder before the other players, or not.

Scientific: cheaper libraries and universities, means you build them fasters, so you the extra research points sooner..this results in a tech rush in the mid game. This can be important for having the technological edge in battle of even for getting sanitation earlier. Also as you build these items quickers, your culture grows earlier. Each free tech, normally save you between 4-6 research turns.

Commerce: I haven't tried this yet, But here is my usual early game situation .. one or two central towns with good production , and all my other town with so much corruption they only produce one shield no matter what territory surrounds them. If commerce allows them to have two shieds, this would double all my production in my outlying towns.

[Fitz]

I think you have to count the affect of the Wonder as well as the shield cost in determining the value of a leader rush. Here's why:

The shield cost is shields used to produce the wonder. If you get it earlier, you can use those shields to build something else instead. Thus, you must count the value of those shields. They are effectively free shields, not shields that would have been payed eventually anyway.

The early affect of the wonder is also not something you would have been getting anyway. If it takes 100 extra turns to build that wonder, you have lost those turns of affect. Once you get the wonder, you have to stop counting this, as the affect will be there anyway (assuming you would have beaten the AI to it). This is part of the value of getting the wonder earlier.

Basically, think of it like this. Assume the Pyramids (ignore incorrect values for it, just think free granaries). I get the wonder right now, as opposed to 100 turns from now (300 shields later). For this price, I get:
1) 300 shields worth of production in that city.
2) 100 turns worth of granary affect (double growth in all 25 cities). Figure out a way to value that pop.

However, you do not include the shield value of a granary, because you would have gotten that anyway when you built the Wonder 100 turns later.

That's not double counting, it's single counting.

[Pyrodrew]

Originally posted by Libertarian
Plus, the workers can quickly build a dynamite infrastructure of roads and railroads that make defensive war a breeze. Zzzzappp! You move your unit from one end to the other instantly to defend key tiles. All the while, these roads are contributing to internal resources (shields, food, etc.).

Workers also can be a great assistance in wartime, literally laying down the roads and railroads as your army advances. They can swarm a newly conquered city and build it up to full capacity in just a few turns.

In a recent game, I was ambushed by the Aztecs. They just attacked out of the blue! Using my workers as a sort of collective tool and my extra income, I was able to crank out a force sufficient not only to repel the Aztecs, but to follow them back into their territory and pillage the hell out of everything they had.

Exactly! Similarly, I was swarmed by Aztecs (3rd largest) & Iroquois (his high-tech sidekick), however his 20 Cavalry & the Iroquois 10 Cavalry coming would overpower my 1-2 Infantry I had guarding my southern roaded cities. So I ditched my plans to take coal from the Zulus & I bought coal from the Russians. Non-industrious workers set up rail in 2 turns... my Americans (Industrious) did it during that turn! There is NO "in 1 turn" set up time for Industrious workers! This makes Industrious workers 3x faster in some cases! Very useful for "need right now" railroads. Result, my railroads were set up so fast, my forces (now a majority) quickly repeled the attack, pushed both Iroquois & Aztecs back, & took some of their cities as well... because of a very efficient & fast built railroad. Leaders, cheap religion, & scouts wouldn't have helped. Less corruption & cheap science may have let me discover Steam Power sooner, but that wouldn't have made me buy the coal from the Russians (if they had it) any sooner. Likewise if I was not attacked I could have focused my industrious workers on building railroads around only my main cities & wonders instead. Thus, Industrious isn't "2nd tier" not only for it's "crank power" as Libertarian & others have stated, but because it provides the flexibility & speed for unexpected surprises that the others cannot.

As far as Early Game - guess who is producing settlers the fastest. Enough said.

I have to agree that Expansionist Bonus is the weakest link tho.

[Dimension]

The biggest problem with trying to equate civ traits directly to shields or gold is that it skews things completely in favor of traits that are better in the late game.

For example, compare Religious to Scientific. By straight comparison, you can see that science buildings cost more than religious buildings, therefore Scientific saves you more shields than Religious. The huge problem is that you generally don't have pop-1 cities building 160-shield Research Labs. By the time you're building Universities and Labs, you've got things rolling pretty well, and you're only saving a couple turns, but (especially on higher difficulties), Temples are essential for even the smallest cities. It's common in the early game to be making settlers and have your cities going back down to 1 population. When those pop-1 cities are producing 3 shields and you decide you need to start making Temples to grow your cities and expand your borders, Religious will make the difference between 15-turn Temples and 30-turn Temples.

Then you can compare the secondary ability of Scientific and Religious. A free tech every era isn't actually a big deal, but the "Statue of Liberty" effect is huge. First, as other people have mentioned, you get the benefit of being able to switch governments when you're at war, instead of trying to make peace to accomodate your government. Besides that, though, you actually do get at least a couple "free techs" just because you're making beakers while another faction would be in anarchy, not to mention shield production.

Forced labor makes Religious even better, if that's possible. Rushing a Temple with 1 population instead of 2... well, that's obviously good. Religious buildings are the ones that need rushed the most. Religious civilizations can also do late-game forced labor by switching governments. If you've got a lot of cities with a lot of food, or if you've got conquered cities with angry citizens who are useless anyway, Religious lets you switch governments, use forced labor, then switch back.

So, directly relating traits to gold/shields has made Scientific look better than Religious, but if you relate the traits in terms of turns saved, Religious is much better than Scientific, and Religious also opens up the option of a "dynamic government" playstyle which is in no way practical with a non-Religious faction.

In SMAC/SMAX, Scientific was a great trait because it gave you an extra free tech at the very beginning of the game, and it increased the speed of your research by 20%. In Civ3, Scientific is a pretty lame trait that doesn't speed up your research at all.

While talking in terms of turns is more practical than talking in terms of gold/shields, it is also a fact that speed in the early game is more important than speed in the late game, which skews things even more in favor of Religious (and Industrious, below).

The gold/shield analysis also made Commercial look twice as good as Industrious, but look at it in terms of turns and when the benefit takes place. Faster workers is more of an early-game benefit, while lower corruption is more of a late-game benefit. Your size-6 city bonuses don't take effect till the mid-game, but you can still say that faster building (extra shield) is more of an early-game benefit, and faster research (extra commerce) is more of a late-game benefit. So, for Industrious, faster workers means less workers, which means more population for settlers, which means faster early expansion.

Whenever you're in a tough situation, think about how different things would be if you had saved a few turns by building your temples a little faster, or building your settlers and roads a little earlier. You'll usually see that things would be much easier. The early game is where speed matters, and it's where the most important decisions are made. I think this makes the Egyptians extremely good, especially for my playstyle, and especially on Emperor/Diety. Obviously, there are a few UU's that are easy to exploit and make other factions stronger for certain playstyles... If you want to get some practice in the late-game, the Greeks are certainly the best for playing a defensive science-driven game. Hoplites a so good, I often don't even need to upgrade them until I start making Modern Armor.

I'm not saying the early game is the only determining factor in your success or failure. If you're a good diplomat, you're relying on the development of other civilizations just as much. Completing your early infrastructure isn't going to change how long it takes you to trade for an important tech, unless you're considering the time it takes to complete roads and how that speeds up moving your scouts around. This is the only argument I can really think of that makes Expansionist sound any good.

[vmxa1]

Dimension, I agree with you on the Sci vs Religous, it is an easy call. The Industrious vs Commerical is a tough one to call. You can make the choice come down on the side of the Civ you want for your style. Commercial and Rel if you want India or Egypt or Rel/Ind. One thing that bugs me is that you may have to take a Civ that has a UU that you would rather not have (no editor). I am not saying it is a bad UU, just say I do not want an early golden age so I would rather not have Immortals as they will win a battle right away. I would like it if you you trigger the GA when you want to after you have earned it. Hoplites are the same way, you get a GA before you can even put four cities together, unless you avoid using them. It negates their value if you can not use them.

[Dimension]

I don't really have too much of a problem with Hoplites causing an early GA. The AI usually isn't dumb enough to attack a fortified Hoplite that would easily win. The Greeks actually have a later GA than most good civs, because Religious or Industrious civs have a real problem with wonder-induced GA's.

Colossus = religious / expansionist
HG = industrious
Oracle = religious
Pyramids = industrious / religious
Great Wall = militaristic / industrious

I'm usually working on Pyramids / Colossus / Oracle by the time I've got my fifth or sixth city down. There isn't much chance of any Religious civ saving their GA for later if you're building wonders.

An early GA is another thing that looks like a VERY bad thing in terms shields/gold, but it's not really that bad when you look at it in terms of turns saved. As long as I've got 8 or 9 cities down by the time Colossus / Pyramids triggers my GA, I'm not disappointed.

I agree that Industrious vs. Commercial depends on style. Industrious is for aggressive players and Commercial is for peaceful players... although the faster workers might make more people lean towards Industrious.

[rah]

While I've loved the discussion, I still have to go with Mil and rel.
Leaders are big ticket and anything that helps gets them is important. Rushing temples for cheap is helpful but ONE TURN ANARCHY is another big ticket. All the others are helpful (except for expansionists) but they're all nickle and dime. I want big ticket items. I changed government over a 12 times in my most recent game. (granted if I hadn't been rel, I would have not done it as often,) but it would be tough for any characteristic to make up for 60 turns of anarchy.

RAH

[Fitz]

Actually rah, I think you're doing that calculation the wrong way. The trait allowed you to switch 12 times instead of 2 times. That's the real value, not the 60 turns of anarchy.

Or maybe it's just too early in the morning for me.

[rah]

Yes there is some truth in that, but I would have had to change at least 6 times anyway. So that's at least 30 turns of anarchy saved. monarchy, rep/maybe, democ, and then switching for 2 or three major wars.

I have problems with prolonged wars in Democ and find it easier to drop into comm. for the blitz. Allowing to rush build those temples and cath to expand culture, harbors to connect to trade networks, and Airports for support troops. Using excessive pop from cities instead of just wasting it while you starve it down, makes too much sense. Then make peace and back to Democ.

RAH
Anyway, it's just a big ticket item that I love having.

[vmxa1]

Speaking of switching Govs, I would like to see a study on their relative values. If you want to war most of the time, why not stay in despot until communism? One switch. If you your prefered form of gov is say republic, how much is it worht to you over say Monarch or Despot or even communist in peace, in war. If you do not know what you give up or what you gain to be in a given form of government, how can you make a determination? What is the cost if you are in a form that your civ dispises? Curently I tend to get to the form that my Civ prefers and avoid the one they don't, but I am not sure if it matters, it should are else why have a preferred or dispised form? I will avoid Dem most of the time as war is problem for it. In civ2 you knew you gave up the boost to research

[Dimension]

Originally posted by rah
While I've loved the discussion, I still have to go with Mil and rel.
Leaders are big ticket and anything that helps gets them is important.
RAH

I think more and more people are starting to agree that Religious is by far the strongest trait, because it has two huge benefits:

1) cheap temples = happy cities and fast culture
2) no anarchy = dynamic govenment for war/peace/forced labor

The other traits depend on playstyle, but I think the major benefit for Industrious makes it better overall. That is, faster workers = less workers and faster roads/lumberjacking. Also very significant is the fact that Religious / Industrial is a huge help in the early game, whereas Expansionist is really the only other trait that helps out early on. A quicker early game gives you the edge for the entire rest of the game.

Commercial and Militaristic are good on paper. Corruption is brutal in the late game, and extra leaders directly translates into more wonders. The problem is that the benefits of Commercial and Militaristic are rather low. Corruption is not reduced much, and your chances of getting a leader isn't raised much. Compare this to Industrious, which gives you a full doubling effect on all workers.

I would have to say that, in a way, Religious does a better job of reducing corruption than Commercial does, because of the easy use of forced labor, the production of which totally ignores corruption. Likewise, Industrious helps you complete wonders faster because you can lumberjack palaces twice as fast.

Still, overall, Commercial is a good choice for the more peaceful, research-driven player looking for space victory in the late game, and Militaristic is no doubt a suitable choice for the ultra-aggressive conqueror. There are also some nice UU's which make weaker traits acceptable.

The real deciding factor for me is that most traits aren't things you miss a lot. I like the Greeks just fine, and the Hoplite is great, but when playing another civ, I don't often find myself saying, "Wow, corruption was a lot lower with the Greeks!" or, "Man, that really hurts having these universities take a couple more turns!" After playing with the Egyptians, though, I do personally find myself in a worse early-game position playing other civs, thinking, "Hey, I could've had another settler but I'm having to build more workers just to get roads done in time," or, "This completely sucks that I can't change governments, because the Romans won't even talk to me so I can end this war, and my Democracy is about to go into massive riots!" Those are things that really do hurt.

[art_vandelai]

Don't forget the fact that all civ's get a worker right from the start. If that worker builds a road and a mine right away, you are that much closer to building that first settler.

E.g. your city has 1 tile of grassland with shield being worked at the start of the game. It will take 10 turns before you gain a pop point, and 5 turns to build a warrior.

An industrious civ will build a road and mine in 5 turns, a non-industrious will build them in 9.

That means 4 extra shields right in the beginning few turns of the game that can increase the speed with which you can build settlers. The roads will increase the speed that your settlers arrive at the places you want to found cities.

Multiply that by the number of turns until your road and mine/irrigation network is completed (usually mid-Industrial era) and you get one heck of a bonus that goes beyond just the shield savings from making fewer workers.

[Arrian]

That, in my opinion, is the major reason to play an industrious civ. The early boost to the land-grab phase. Quicker mines and roads around the capitol, and then roads out toward city sites. That can seriously speed up expansion.

-Arrian

[Molotov99]

I am very fond of expansionist civs. Having the scout so early in the game i find very useful in setting my early city sites, for contacting other civilizations and finding most of the goodie huts. How would you factor the extra gold/shield values obtained by being the first civ to find all the others thereby trading your way to more and more advances and making cash from trading communications. Also by finding prime city sites early you can rush your settlers into place. Pair it with any of the others, personally I like commerce so play England.

Thoughts from sunny England
Rich

[Lucky_Shot]

This string begun by NATO's excellent innovation of a quantitative analysis of Civ traits has been stimulating reading. One very important consideration, though, is that of "compound interest." Civ is a game of growth, much like a retirement account. Early advantages in commerce or industry, along with others, must be thought of in terms of adding to a civilization's "principle" which grows at some rate of "compound interest. These contributions add to principle both directly (as has been calculated) and through their effects on growth of total holdings (improvements, population growth, etc.) The effects are truly that of compounding because early investments, whatever one's playing style, have profound effects on not only on the end total but also the rate of growth. This notion seems to have been neglected throughout the discussion.

Because Civ can be boiled down to a race to grow, albeit with combat and other factors added in, the total contribution of any advantage must be thought of in terms of not only simple addition to principle, but also how much that advantage is both compounded by growth and, indeed, serves to accelerate the rate of growth itself. The way it is being calculated now, Commerce or Industry (or whatever trait advantage) are treated as though they merely added to principle over the course of the game. Were I to perform a similar calculation for my retirement fund, I could count on nothing more than the total of my personal contributions, a grim picture indeed. Instead I assume that contributions are increased by some rate of interest, and money I contribute earlier will dramatically increase the account value at retirement. Now if only I could find a mutual fund that grew like my citizens (and their works) in Civilization!

Consideration of this growth phenomenon is indeed implicit in giving much value at all to the Expansionist trait--two fast moving units on day one are like contributing a big chunk of change to one's retirement on the first day on the job. A cardinal principle of investment is that small contributions early are better than big contributions late.

NATO has made a groundbreaking suggestion with the idea of quantifying the effects of traits. Adding in consideration of the benefits under an investment model with compounding interest (based on some rational rate of observed growth), would seem to offer a much better sense of which traits offer the greatest game-long advantage. I imagine the totals could be quite remarkable.

I must add an admission that not only do I lack the mathematical skills to carry out my ambitious suggestion, I have yet to actually play Civ III. I have been an all too avid player of I and II and, honestly, am afraid to buy a copy of III. I have been satisfying my curiosity (for now) by reading about it. It sounds wonderful.

[Dimension]

Lucky_Shot: Actually, while nobody has referred to it as "compound interest," the fact that early-game development is undervalued in nato's initial analysis has been discussed at length.

Originally posted by rah
While I've loved the discussion, I still have to go with Mil and rel.
Leaders are big ticket and anything that helps gets them is important.

I definitely agree that Religious is the best civ trait, but Industrious contributes so much to early development that it's very hard to beat. The problem I'm running into is that I'm adopting a more and more aggressive playstyle, and (like many others) have found that captured workers are the way to go. That's too bad, because in the midgame when you're using nothing but captured workers, you've nearly made the Industrious trait useless. That extra 1 shield for big cities isn't exactly a big deal.

This thinking has lead me to look for something better than Industrious when playing aggressively, which could only be Militaristic. The Japanese & Aztecs are both Militaristic/Religious, but you have to go with the Aztecs, because they've got a UU that makes Expansionist look like a joke. If there's anything that can compensate for not being Industrious, it's stealing workers with Jaguar Warriors.

Militaristic is also great in the early game, because you can also build fast barracks and make archers from turn 1, and the increased chance of leaders (no matter how small) can make a huge difference on Deity when wonders are hard to get.

I still think I would rate the Egyptians as the best overall civ, but the Greeks are probably the best for peaceful players on lower difficulties, and the Aztecs might be the best for aggressive players on higher difficulties.

[RamboMilitia]

I'll add my 2 cents:
About religion. It might be argued that the cheaper temples and especially cathedrals has another advantage not yet accounted for: Less civil disorder and thus more optimal use of production/commerce.
Also I feel it might be interesting to quanify in terms of culture instead of gold. Although harder to qualify it is my feeling that whatever way one plays this is the only deciding factor.

[godinex]

In a Huge map the best combination is Industrious and comercial, but in a small one ?

[Arrian]

Actually, especially with the 1.17 patch, one could argue that Expansionist/anything is best for a huge map. The Iroquois are rather appealing on a map like that.

-Arrian

[Dis]

I'm still pissed so many play the sorry french.

You guys are too peaceful.

I'm all for leaving in a Theocracy . Military and Religious all the way for me. I crush the non-believers .

Religious- you have to have temples to help thwart city flipping. So you are going to build them anyways.

Military. Well this is only useful if you like war. And I do. But to wage war without militaristic trait just isn't that fun.

[Analyst Redux]

Originally posted by Arrian
Actually, especially with the 1.17 patch, one could argue that Expansionist/anything is best for a huge map. The Iroquois are rather appealing on a map like that.

-Arrian

Yes, a concept that has managed to go unmentioned throughout this thread is that the worldmap has value, and in the ancient era, the player with the most complete worldmap is often the player with the single item of greatest value (as calculated by the AI for trade purposes). Soren's comments about AIs leveraging the value of a more complete worldmap in the game, post-1.17 patch, are an important cue that this value-added feature of being an expansionist civ shouldn't be underestimated. It makes a world of difference in the flow of the whole game to cross over from the ancient to midieval eras with tech parity, rather than still struggling to close the gap.

Like other expansionist civ values, it is pretty much run its course by the end of the ancient era. But as other posters point out, value added in the ancient era pyramids, meaning that though the acts of expansion end, the effects are felt throughout the game.